[checkmk-commits] Check_MK Git: check_mk: Check manpages for Veritas and LVM checks

Module: check_mk Branch: master Commit: 8036445050d7070e2257ff019c833bffd13e1aa0 URL: http://git.mathias-kettner.de/git/?p=check_mk.git;a=commit;h=8036445050d707… Author: Florian Heigl &lt;fh(a)mathias-kettner.de&gt; Date: Mon Jun 4 11:33:17 2012 +0200 Check manpages for Veritas and LVM checks --- checkman/lvm_volstats | 13 +++++++++++++ checkman/vxvm_enclosures | 21 +++++++++++++++++++++ checkman/vxvm_multipath | 18 ++++++++++++++++++ checkman/vxvm_objstatus | 24 ++++++++++++++++++++++++ checkman/vxvm_volstats | 13 +++++++++++++ checks/vxvm_volstats | 4 ---- 6 files changed, 89 insertions(+), 4 deletions(-) diff --git a/checkman/lvm_volstats b/checkman/lvm_volstats new file mode 100644 index 0000000..0a551e6 --- /dev/null +++ b/checkman/lvm_volstats @@ -0,0 +1,13 @@ +title: Linux LVM Volume statistics +agents: linux +author: Bastian Kuhn &lt;bk(a)mathias-kettner.de&gt; +license: GPL +distribution: check_mk +description: + This check traces IO activity on volumes created in Linux LVM + It reads the same counters as the basic {diskstat} check, but + limits the results on per-Volume basis. + +inventory: + One item is generated per volume in the volume manager. + The naming is {volumegroup}-{volume}. diff --git a/checkman/vxvm_enclosures b/checkman/vxvm_enclosures new file mode 100644 index 0000000..da0ddc1 --- /dev/null +++ b/checkman/vxvm_enclosures @@ -0,0 +1,21 @@ +title: VxVM Enclosures +agents: hpux, linux +author: Bastian Kuhn &lt;bk(a)mathias-kettner.de&gt; +license: GPL +distribution: check_mk +description: + This check verifies status of disk enclosures or arrays + connected to a host running Veritas VxVM. + + It is in status {OK} an enclosure is in status CONNECTED. + Any other status will be reported as a {CRIT} status, assuming + the enclosure has lost connectivity or has been removed. + Veritas also lists the local disks of a system in a + "pseudo-enclosure" named {OTHER_DISKS}. If you do not run + the local disks under Veritas control, you can add these to + {ignored_services}. + +inventory: + One item is generated per VxVM disk enclosure, the naming is + inherited from VxVM. See the man page for {vxdmpadm} about + naming enclosures. diff --git a/checkman/vxvm_multipath b/checkman/vxvm_multipath new file mode 100644 index 0000000..c00656d --- /dev/null +++ b/checkman/vxvm_multipath @@ -0,0 +1,18 @@ +title: VxVM Multipath +agents: hpux, linux +author: Bastian Kuhn &lt;bk(a)mathias-kettner.de&gt; +license: GPL +distribution: check_mk +description: + This check verifies status of multipath devices managed using + Veritas native multipathing {VxDMP} on a host running Veritas VxVM. + It will return {OK} if all configured paths are active. + Otherwise, it will decide based on the number of active paths. + If more than 50% of the paths are still active, it will return a + {WARN} status, or a {CRIT} if less than 50% - or none - paths are + still working. + + +inventory: + One item is generated per multipathed LUN. It is named after the + volume managers' {access name} for the disk. diff --git a/checkman/vxvm_objstatus b/checkman/vxvm_objstatus new file mode 100644 index 0000000..557f488 --- /dev/null +++ b/checkman/vxvm_objstatus @@ -0,0 +1,24 @@ +title: status of objects in VxVM +agents: hpux, linux +author: Bastian Kuhn &lt;bk(a)mathias-kettner.de&gt; +license: GPL +distribution: check_mk +description: + This check is used to track the status of objects in Veritas VxVM. + Currently, it can only track the status of {v} - Volume level objects. + It will be {OK} if a volume is an active status, or if it was + dismounted/stopped cleanly by admin action. If any other status is + detected, it will go to {CRIT}. + + If a diskgroup goes to {disabled} by catastrophic failure or admin + error, the check can no longer report the volume status, as the + volumes as objects become unavailable. + + The check should be extensible for tracking status in multiple layers. + + +inventory: + One item is generated per volume in the volume manager. + The naming is {diskgroup}-{volume}. + The inventory includes layered volumes {v2}, if you do not need data on + these, use {ignored_services} to exclude them from monitoring. diff --git a/checkman/vxvm_volstats b/checkman/vxvm_volstats new file mode 100644 index 0000000..b7c193e --- /dev/null +++ b/checkman/vxvm_volstats @@ -0,0 +1,13 @@ +title: VxVM Volume statistics +agents: hpux, linux +author: Bastian Kuhn &lt;bk(a)mathias-kettner.de&gt; +license: GPL +distribution: check_mk +description: + This check traces IO activity on volumes created by Veritas VxVM. + It uses the {vxstat} utility to track per-KB statistics. + + +inventory: + One item is generated per volume in the volume manager. + The naming is {diskgroup}-{volume}. diff --git a/checks/vxvm_volstats b/checks/vxvm_volstats index f32a68f..1d4eb3a 100644 --- a/checks/vxvm_volstats +++ b/checks/vxvm_volstats @@ -24,10 +24,6 @@ # to the Free Software Foundation, Inc., 51 Franklin St, Fifth Floor, # Boston, MA 02110-1301 USA. -# <<<aix_perflib_hbastat>>> -# Convert to generic format needed for diskstat.include. Upper example -# would look like: -# [ ( 'sissas0', 204494394, 156524126 ), ('fcs3', 1760, 0), ... ] def convert_vxvm_volstats(info): return [ ( l[0], int(l[1])*2, int(l[2])*2) for l in info[1:] ]